Abstract: This is a revised proposal in response to PA-18-141 and the NIH HEAL initiative, designed to unravel mechanisms that underlie chronic osteoarthritis (OA) knee pain. OA is the leading musculoskeletal chronic pain condition worldwide, yet little is known about the mechanisms of chronic OA pain, reflected in the fact that current pharmacologic approaches are minimally effective and new treatments have not been developed. In contrast, joint replacement surgery is highly effective in most, but not all, patients with OA. For unknown reasons, around 20% (>140,000 cases in 2017 in the US alone) of OA knee replacement surgeries (TKR) fail to relieve pain. We and others have shown that in people with chronic OA pain, the brain shows maladaptive reorganization of the neocortex, diminished volumes of sub-cortical limbic structures, distinct brain activity for OA pain, and global disruption of functional information integration. Together these results imply altered personality, psychosocial status, and abnormalities in abilities for cognition, emotion, sensation and motor function (CESM-abilities), which to our knowledge remain essentially unexplored in OA. In addition, nociceptive processes (peripheral and central sensitization, descending modulation) have been considered as possibly being important for chronicity of OA. Hence, the primary goals of this proposal are (1) to characterize the neurologic mechanisms for chronic OA knee pain, and (2) to define neurologic mechanisms that differentiate success and failure of TKR. We propose testable hypotheses regarding mechanisms underlying chronic OA pain and those that control TKR outcomes. In Aim 1, we will study a large group of OA pain patients prior to TKR, as well as OA pain patients not undergoing TKR (positive control) and healthy individuals (negative control), to characterize brain circuitries (T1, DMRI, resting state fMRI) and determine how these map to nociception, to pain and related psychosocial status, personality, and CESM-abilities. Since ~80% of TKR are successful in the long term (12 months), we hypothesize that in these cases, the dominant parameter controlling pain is the OA joint-related nociceptive processes; while in cases where TKR fails in the long term, there is a stronger dependence on psychosocial attributes and personality (based on limbic brain properties). The latter hypothesis will be tested both over the short term (3 months post-TKR in Aim 2A) and in the long term (12 months post-TKR in Aim 2B), by constructing models from pre-TKR measures (collected in aim 1) to predict knee pain in the short and long term after TKR. In Aim 3, subgroups of patients with the greatest and least pain relief at 3 months post-TKR will be fully reassessed for outcomes deemed relevant (in Aim 1), followed, and then reassessed again at 12 months post-TKR. Outcome contrasts between groups, and within groups in time, will allow us to identify consequences of knee surgery. These outlined studies expand on our current knowledge regarding mechanisms of chronic pain in general, and more specifically for OA and for post-TKR pain, potentially unraveling novel therapeutic targets.